Recycling of Discarded Mattresses Through Extended Producer Responsibility: Is It More Cost-Effective than Incineration?

Abstract

About half of the discarded mattresses in the Netherlands are recycled, and the other half are incinerated. Based on a recently implemented sustainability measure of extended producer responsibility, the recycling rate should increase to 75% in 2028. Thereby, a recycling fee of EUR 12.5 per mattress should be introduced to finance the infrastructure to increase recycling. This study investigates the potential cost-effectiveness of mattress recycling through the reuse of materials, compared to the incineration of mattresses in Dutch energy-to-waste plants. The benefits of recycling include the avoidance of CO₂ that would otherwise be released by incineration and the displacement of virgin material that would otherwise be used by producers as raw material. However, there are also significant costs associated with the collection and recycling process of complex products such as mattresses. Taking all factors into account, the cost of saving one ton of CO₂ through mattress recycling is EUR 138. This is higher than alternatives such as wind energy, ETS prices, or carbon capture and storage. If the replacement energy is fully CO₂-neutral or the recycling fee is lower, the costs of one ton of CO₂ decrease but are still higher than most alternatives.

1. Introduction

1.1. Waste Recycling or Incineration

In the Netherlands, essentially only waste for which no recycling or incineration option exists is used as landfill [1]. Dutch municipalities recycled 57% of their waste in 2020. The rest of Dutch municipal waste is incinerated in efficient waste-to-energy plants [2]. The European Union (EU) establishes a goal of recycling 65% of municipal waste by 2030. Nevertheless, the Dutch recycling rate is far above the EU average of 48% [3]. Moreover, the Netherlands has a strict recycling target by demanding that every citizen should be allowed to have no more than one-hundred kg of unsorted waste per year in 2025 [1]. Therefore, the Netherlands explores a narrow interpretation of the circular economy (CE). In an analysis of 114 definitions for CE, Kirchherr et al. (2017) mention that CE would be about reduction, reuse, recycling, and recovery, but that many policies, and especially the Dutch policy, have been oriented towards promoting recycling [4]. In the past decade, it has been the Dutch administration’s policy to implement separate curb side collection of paper, glass, plastics, and compostable waste and to ensure the provision of special containers in most municipalities, although the effectiveness in increasing the recycling rate is small [5]. Despite different policy measures to increase sustainability and the circular economy, there has been little attention to compare these Dutch measures in terms of environmental sustainability and economic efficiency.

In 2018, in the Netherlands, 52% of plastic (packaging) waste was registered as recycled, although this yield has been disputed as only approximately half is usable in the main recycled products [6]. In addition to this, Gradus et al. (2017) show plastic waste separation by residents is very expensive [2]. Municipalities must set up an infrastructure for plastic collection from residential properties. Therefore, the cost-effectiveness of plastic recycling—in terms of costs per ton (1000 kg) of carbon dioxide (CO₂) reduction—is low. Based on Dutch data, a saving of 1 t of CO₂ through plastic recycling compared with incineration costs approximately EUR 180 [2], which was substantially more than the Emissions Trading System (ETS) prices at that time, as well as current ETS prices (approximately EUR 75 at the end of 2024). Alternatives such as wind energy (approx. EUR 30) or carbon capture and storage (CCS) in the old natural gas fields in the North Sea (approx. EUR 85) are much cheaper.

In addition, for (packaging) glass the EU advocates a recycling rate of 75%. In 2017, the Netherlands recycled 86% of their (packaging) glass, far above the EU goal [7]. Nevertheless, the Netherlands has set an even higher goal by demanding that 90% should be recycled in 2021, and the Human Environment & Transport Inspectorate (ILT) forced the Packaging Waste Fund to reach this goal by increasing the number of bottle banks by 800. In [7] an effectiveness analysis is given and based on an empirical analysis there is an increase of only 0.07% if the number of bottle banks increases by 800 and, to achieve the 90% target, the number of bottle banks should increase by 43,000, nearly three times the current infrastructure. The Packaging Waste Fund uses EUR 7000 as a proxy for costs of installing a bottle bank and, therefore, total costs will increase by EUR 300 million.

1.2. Mattress Recycling

Importantly, 30% of Dutch household waste are bulk waste such as mattresses, construction waste, flat glass, and demolition waste or floor coverings. Dutch municipalities should have a facility where citizens can bring these different waste streams so waste can be separated [8]. Mattresses are bulk waste, where a special target is set by the government. In 2028, the proportion of mattresses in the Netherlands that are recycled should be at least 75% of the number of mattresses placed on the market. This is combined with the manufacturer’s responsibility for mattresses, which is mandatory in the Netherlands since 2022 [9]. Extended producer responsibility (EPR) means that producers or importers are (co-)responsible for the waste management of the products they sell. This principle involves a shared chain responsibility because other links in the chain (such as municipalities and retailers) also take responsibility. Section 2 contains more details about the Dutch system. Nevertheless, recycling of mattresses is a challenge as it is costly to dismantle them and the secondhand materials, in general, have low revenues. Therefore, the environmental impact and costs of alternatives such as incineration are important.

In some US states there are some efforts to stimulate mattress recycling. In California, Connecticut, and Rhode Island, mattress recycling laws have been passed [10]. State law requires retailers to collect a recycling fee on each mattress that is sold. Nevertheless, in the United States, the great majority of mattresses and box springs also end up in landfill [10]. Incineration is not an often used option in the US. Yamamoto and Kinnaman (2022) notice that when the ‘‘Renewable Energy Facility’’ opened in June of 2015 in West Palm Beach, Florida, it marked the first time in over 20 years that a new solid waste incinerator began operations in the United States [11]. Geyer et al. (2016) calculated the greenhouse gas savings of recycling mattresses and compared this with landfilling, where the energy content is only partially preserved (compared to incineration in waste-to-energy plants), and found that recycling is superior [10]. The most recent available EU data on mattress reuse from 2014 suggests that 60% of discarded mattresses thrown out are sent to landfill and the other 40% are incinerated [12]. At the same time, the increase in the demand for energy due to economic growth and the decrease in landfill space in almost all developed countries makes pyrolysis or incineration processes desirable routes for polyurethane (PU) foam from discarded mattresses, allowing for the production of fuels, gases, and energy [12,13].

In the Netherlands, approximately 1.5 million mattresses are discarded each year, of which 1.2 million are household mattresses [14]. Although more and more effort goes into recycling, a cost effectiveness analysis of incineration versus recycling mattresses is lacking, and this paper tries to fill this gap by estimating the costs and revenues of both options. In environmental economics a cost effectiveness calculation of the most common treatment methods is quite common (see, for example, for removing microplastics in wastewater [15]). The main benefit from mattress recycling is the avoidance of CO₂ emissions that otherwise would occur during incineration. At the same time, there are significant costs involved with the collection and recycling of mattresses. Importantly, municipalities should invest in a system to store these waste mattresses, otherwise they become wet and polluted, making them unsuited for recycling. The benefit from mattress incineration is the energy that can be recovered, which reduces emissions in the regular energy production sector. In the robustness analysis we also discuss the consequences of if there is a fully neutral energy sector.

1.3. Aims and Structure of This Article

The objective of this study is to assess the economic and environmental feasibility of mattress recycling compared to incineration in the Netherlands, within the framework of extended producer responsibility and in light of national recycling targets set for 2028. This article thus addresses a gap in the literature by providing a cost effectiveness assessment of mattress recycling and incineration with energy recovery in the Netherlands, focusing on environmental impacts and economic costs. The main cost associated with incineration is that this requires a waste-to-energy plant with the associated capital investments. Summing up the costs and revenues from both waste treatment options and comparing the results will lead to an implicit CO₂ abatement price of 138 EUR/t of CO₂ in the case of recycling. We contrast this implicit CO₂ price with other prices and workable alternatives (e.g., renewable energy or CO₂ capture and sequestration (CCS)). In the literature more and more studies are available with the cost of reducing CO₂ emissions of different policy measures [16]. We demonstrate that the implicit price of mattress recycling vs. incineration is higher than past and present ETS prices, external costs, and other options, such as the generation of solar or wind energy. In addition, the social costs of carbon dioxide emissions, even at an average of EUR 50 per tonne CO₂, are still significantly less than the shadow CO₂ price for recycling mattresses [16]. To test the validity of our conclusions we include a sensitivity analysis.

The rest of this article is structured as follows. The producer liability for mattresses in the Netherlands is described in Section 2. The cost effectiveness analysis is presented in Section 3, where we first discuss the methodology and data we used before proceeding to the analysis itself (Section 4.1). The sensitivity analysis is presented in Section 4.2, and Section 5 and Section 6 conclude, discuss policy implications, and offer ideas for additional research.

2. Research Context

EPR is a widely used policy to support the transition to a circular economy. In the USA, EPR is implemented, but on a state level and in so-called blue states such as California, Rhode Island, and Connecticut (e.g., [17]). EPR schemes for batteries, end-of-life vehicles, electric and electronic equipment, and packaging are implemented across the EU, but for mattresses an indicative timeline for adoption has been set at 2029 [18]. In the Netherlands, EPR also applies to car tires, paper and cardboard, and flat glass [19]. As of 1 January 2022, mattress manufacturers and importers are obliged to pay a disposal and recycling fee on all mattresses [9]. This fee is not separately presented on the bill for consumers. Notably, this legislation is an agreement between the State of the Netherlands and the Mattress Recycling Netherlands Foundation (MRN), a foundation consisting of five Dutch mattress manufacturers and importers (IKEA B.V., Beter Bed Holding N.V., Koninklijke Auping B.V., Swiss Sense B.V., and Hilding Anders Netherlands B.V.) that make up 86% of the market share and took the initiative to collectively manage mattress recycling. In addition, EPR for mattresses is implemented through Generally Binding Declarations. The EPR for mattresses started on a voluntary basis and is financed by contributions made by a group of producers. An AVV is a ministerial order, by which an agreement on the financial contribution for post-consumer collection, sorting, and treatment is declared binding for all entities putting a product on the market. Due to this legislation, MRN is charged with the following:

• Determining the amount of a recycling fee to be added to the mattress price.
• Setting up a collection and recycling infrastructure for municipalities and stores funded by the fee.
• Establishing, and periodically reviewing and revising, performance measures, such as a yearly increase in recycling by 5%, such that a recycling rate of 75% will be reached in 2028.

Each producer or store must declare the volume of mattresses they sell. To avoid administrative costs, small stores (with fewer than two hundred mattresses yearly) have been exempted from this obligation. The expectation is that in 2028, 1.6 million mattresses will be sold (and if 75% recycling is achieved, 1.2 million mattresses will be recycled).

In the Netherlands, all costs for the collection, transport, and recycling of mattresses will be fully reimbursed by MRN in 2028. To reach the 75% goal, municipalities should encourage consumers to bring end-of-life mattresses to municipal dumps. Nowadays, almost all Dutch municipalities have facilities for bulky household waste such as mattresses [8]. These centers have facilities for collecting different waste streams, such as chemical waste (batteries, medicine, solvents, paint, etc.), construction and demolition waste, and flat glass. Batteries can be handed in at shops and supermarkets as well. Some municipalities charge extra for this bulky waste, which can be a disincentive to bring mattresses to these dumps.

Moreover, the authors of [20] estimate that to achieve this recycling goal by 2028 a fee of EUR 12.50 for a mattress should be maintained. Most recently, MRN reports a recycling fee of EUR 8.65 (excluding VAT) and EUR 10.47 (including VAT) in 2028, but according to the Dutch council for municipality cleaning (NVRD) this is not sufficient to reimburse all costs to municipalities [9]. Due to a lack of finalized data we work on the assumption of a fee of EUR 12.50, as used in [20], and test the sensitivity for other assumptions as well. Notably, if four (out of five) companies agree and the market share is at least 66%, MRN can increase this fee, for example, due to increased transport costs, and can also differentiate based on differences in costs of companies, for example, due to the sales of recycled material [20].

3. Method

3.1. Mattress Characteristics and Recycling Parameters

Mattresses are composed of various materials that must be separated in a complex treatment process before they can be recycled (see, e.g., [13,21]). Assuming clean and dry mattresses, recycling of up to 90% of the mattresses’ weight is possible in practice, thus leading to a very small residue that must be incinerated [9]. However, only clean and dry mattresses can be recycled; mattresses that are not dry or not clean should be considered as residual waste and must be incinerated.

Dutch waste-to-energy plants use state-of-the-art technology, which filters out most air pollutants such as SO₂ and NO_x [2]. This is the result of additional national environmental standards, which go beyond the European minimum requirements. In their comparison of Dutch incineration versus landfill, Ref. [22] demonstrates that, given the technology of the 1990s, CO₂ dominates the environmental cost for air emissions, accounting for 90% of the total environmental cost. Therefore, in this cost-effectiveness analysis the environmental impact is expressed in terms of CO₂. A waste-to-energy plant calculates a gate fee for each input. This rate reflects the energy yield of the input material.

To assess the relative cost-effectiveness of mattress recycling, the specific revenues and costs of recycling versus incineration are compared. In

Table 1. Data and sources for the cost-effectiveness analysis.

Input	Figure	Source
Mattress characteristics
Energy content mattress	30.0 MJ/kg	[23]
Waste incineration parameters
Collection costs mixed waste	EUR 70/t	[2]
Incineration costs mixed waste	EUR 125/t	[2]
Recycling parameters
Mattress recovery percentage	90%	[9]
Recycling fee municipalities	EUR 375/t	[20]
Post-collection	EUR 250/t	[20]
Market costs recycling PU foam	EUR −34/t	[20]
Market costs bonded PU foam	EUR 50/t	[20]
Market costs spring steel	EUR 1100/t	LME: average price 2020
Composition of average mattress
PU Foam	40%	[24]
Ticking	25%	[24]
Spring steel	20%	[24]
Latex	10%	[24]
Residual materials (e.g., plastics)	5%	[24]
New product	1040 kg CO2/t	[25]
Production PE for new mattress	−1200 kg CO2/t	[25]
CO2 emissions incineration	−2666 kg CO2/t	[25]
Electricity yield plant	1085 kg CO2/t	[25]
Heat output plant	170 kg CO2/t	[25]
Transport parameters
Distance to incinerator/sorting	100 km	Assumption: 50 km twice
CO2 road transportation	0.073 kg/km t	[26]
Road transport costs	EUR 0.043/km t	[26]
Price parameters
Electricity price	EUR 98.67/MWh	Based on average cal. 2021 APX prices
Heat and steam price	EUR 14.8/MWh	[27]

the main publicly available data for the cost-effectiveness analysis and their sources are specified.

We base our estimation of the costs of collecting, transporting, and recycling mattresses in the Netherlands on figures from a single Dutch mattress company [20]. Collectors of mattresses, such as municipalities and waste management companies and store owners or retailers, can receive this fee. The cost data are derived from one company and may not reflect the full range of company sizes or regional variations within the mattress sector. Therefore, in the sensitivity analysis, other costs are analyzed as well.

Mattresses have two important characteristics that make their recycling complex [21]. The first one is their high volume that is a disadvantage in the transport of these products to the recycling plant and the second characteristic is their heterogeneous composition, with materials such as PU foam, metals, and textiles, that requires dismantling prior to recycling.

A standard mattress consists of three main components: a core, a covering layer, and ticking [28]. The core of mattresses consists mostly of PU foam (average 40%), which is made through a polymerization process. These figures are based on averages. It should be mentioned that mattresses may vary considerably in terms of the composition (which may also change depending on the available technologies or consumer trends in the market). In addition, ticking (average 25%) and steel (average 20%) have an approximately equal share in an average mattress. The ticking consists of a cotton layer, a fleece so that the filling cannot come out, and a cotton cloth for finishing. The remaining part is latex (10%) and residuals such as glue (the remaining 5%), which, in general, are more difficult to recycle. We assume that the mattresses’ energy content is purchased at market rates. Mattresses have, in general, an energy balance of 30 MJ/kg [28], which produces three times as much energy as when mixed waste is incinerated [23]. Therefore, a decrease in mattress waste in mixed waste could lower the energy output per unit of input and thus increase overall incineration costs (by reducing effectiveness and energy return). Following [29] we take it to be 20 kg per mattress. If mattresses are incinerated, this means that more virgin material is needed to produce new mattresses. Like [2], these are presented as the opportunity costs of incineration. To allow a comparison with recycling, the calculations are based on one ton of mattresses and assumes they are now produced from virgin material.

Based on a discussion with an expert [30], the treatment and incineration costs of a ton of (contaminated) mattresses are estimated at between EUR 100 and EUR 150 due to the heterogeneous composition and the relatively labor-intensive sorting process (see also [17]). Various figures and assumptions circulate in the Netherlands about the cost of collecting, transporting, and recycling mattresses [14]. Due to lack of optimal data, we make use of the assumptions used in the EPR. Based on figures from a Dutch mattress company [20], it is estimated that to achieve the goal of 75% a fee of EUR 12.50 for a mattress weighing 20 kg on average should be maintained. We consider this figure to be the best available proxy of costs. Since this approximation is an estimate of costs, the sensitivity of the results is also tested for a change in fees. For the revenue side, the material composition of the average mattress as described by [24] is assumed. The assumption for the treatment of PU foam is that one half of the foam can be re-used (i.e., high quality recycling), and the other half must be incinerated. In addition, average scrap prices for old stainless steel from the London Metal Exchange over the year 2020 are used to provide an indicative price for spring steel from mattresses not designed for re-use. Revenues from secondary textiles, which among others are recycled mostly into wiping cloths [14], are relatively small and are therefore not considered in this analysis.

3.2. Opportunity Costs

If mattresses are recycled, less energy is produced by waste-to-energy plants. Our calculations are based on the same energy output produced through regular electricity generation to make this comparable to incineration. These are described as the recycling opportunity costs. The waste-to-energy plant’s total costs, including operating and capital expenses, are calculated using publicly available data. These costs are based on the integral cost price of waste incineration, as given in [2].

In the past, Ref. [25] commissioned research into the environmental benefits of recycling and incineration. This showed that recycling has higher yields in CO₂ than incineration; according to their calculations, the amount of CO₂ saved by recycling mattresses of high quality is 23.25 kg of CO₂ per mattress. If we assume 20 kg per mattress, this will amount to 1.16 tons of CO₂ per ton of recycled mattresses. Compared to the case of plastic waste, which has a similar calorific value, this outcome is plausible (see also [2]). We assume an average distance of 50 km to the incinerator or sorting installation. Although the Netherlands is a relatively small country with well-developed waste management infrastructure, the number of mattress recycling installations is limited. Furthermore, Ref. [26] considers transportation costs and the opportunity costs of electricity generation from the waste-to-energy plant. Finally, the CO₂ yields for electricity and heat are estimated based on available average yields of Dutch waste-to-energy plants, as indicated by [27].

4. Results

4.1. Cost-Effectiveness Analysis

Table 2. Recycling and incineration of one ton of discarded mattresses: net cost in EUR/t.

	Recycling	Incineration
Remuneration costs	375	74
Processing	250	125
Revenues	−203	−124
Subtotal	422	75
Opportunity costs of energy	112
Opportunity costs of secondary material		234
Total	533	309

gives the costs and revenues of one ton of (discarded) mattresses for recycling and for recovering energy from the components of mattresses in a waste energy plant.

First, the costs of collection and processing are considered. For mattress incineration, the collection costs (i.e., expenses for moving the mattresses from the collection point to the disposal or processing facility) are 70 EUR/t, and the transportation costs (i.e., expenses for gathering mattresses from its source to a collection point) are 4 EUR/t, which together amount to 74 EUR/t. For mattress recycling the remuneration costs are 375 EUR/t, with EUR 325 reserved for logistics, including setting up a municipal drop-off point for mattresses, and EUR 50 benefiting MRN [20]. Consumers themselves should deliver the mattresses to a store or local authority.

Second, the collection and transport costs for mattresses are significantly higher than for mixed waste. This is because the density of mattresses is considerably lower than that of mixed waste (see also [14]). In other words, mattresses require more transport costs per ton than mixed waste streams with a higher density. Second, the processing costs of mattress recycling are significantly higher than those of incineration, caused by the greater requirement of careful disassembly due to the separate components. In the case of mattress recycling, we consider the proceeds from the sale of secondary materials, and in the case of mattress incineration, we consider energy costs. The processing cost for the recycling option is 250 EUR/t of mattresses (assuming the recycling of PU foam and spring steel), while the processing cost for the incineration option is 125 EUR/t. Selling PU foam and spring steel generated EUR 203 in revenue. The costs and revenues associated with incineration are calculated using incineration costs of EUR 125 and the average 2021 revenues from electricity and heat production of EUR 83 and EUR 41, respectively. In summary, the total (monetary) cost is 422 EUR/t for the mattress recycling option, while the cost of energy recovery from mattresses is 75 EUR/t.

Third, we consider the opportunity cost of recycling mattresses in the case of mattress incineration, and conversely the opportunity cost of energy recovery from mattresses in the case of mattress recycling (see also [2]). In other words, there is an energy deficit that needs to be made up when used mattresses are separated and recycled at the source rather than being burned. The energy loss in the case of mattress recycling is equal to 112 EUR/t—the equivalent energy value of recycled mattresses (90% of incineration of one ton mattresses) that would otherwise have been used for energy recovery. On the other hand, there is a shortage of secondary materials that must be considered when mattresses are not recycled but instead burned. When mattresses are incinerated, there is a deficit in the volume of secondary materials (recycled mattresses). The value of recycled mattresses lost to incineration is 234 EUR/t.

In conclusion, the net cost of recycling one ton of mattresses is much higher than the cost of recovering energy from them. There is a difference of 224 EUR/t between the total cost of recycling mattresses (533 EUR/t) and the cost of recovering energy from mattresses (309 EUR/t). The CO₂ emissions of both options are shown in

Table 3. Recycling and incineration: CO₂ emissions in tons of CO₂ per ton of mattress waste.

	Recycling	Incineration
Energy recovery	0.27	2.67
Recycling	0.16
Transport	0.02	0.01
Subtotal	0.45	2.67
Opportunity emissions of energy	0.76
Opportunity emissions of mattress waste		0.16
Total	1.21	2.83

.

One ton of mattresses produces 2.67 tons of CO₂ given current Dutch incineration technology. Since 10% of separated mattresses are eventually incinerated, separating one ton of mattresses via energy recovery yields 0.27 tons of CO₂. The recycling of mattresses is a sophisticated industrial process that also releases CO₂, mainly through energy consumption. Based on previous research [25], we assume that recycling 0.9 tons of mattresses produces 0.16 tons of CO₂. Another important point is that dismantling can cause hygiene and health-related problems due to bacteria content, sweat, and dust, but these drawbacks can be avoided with a thermal process. According to the National Institute for Public Health and the Environment a thermal process of 100 °C is sufficient for complete disinfection and consequently there should be an investigation into whether the current processes, where some use 60 °C, should be upgraded to a higher temperature and therefore use more energy [31]. Furthermore, mattresses transportation leads to some CO₂ emissions, although these emissions are moderate.

We estimate the opportunity emissions between the two options, along with the lost income or expense from choosing between different options for mattress processing. The opportunity emissions are calculated using the most popular alternative production method for either energy production or the production of recycled mattress materials. This is based on the average energy mix currently used in the Netherlands, which includes both conventional and renewable sources of electricity generation [28]. The most popular alternative method for producing heat is the use of gas turbines. Since no other process can produce this quality of material reuse, we assume that the alternative emissions for secondary materials are those typically associated with mattress recycling on an industrial scale.

According to Table 3, the energy recovery option’s CO₂ emissions are 1.62 tons higher than those of the recycling option. The higher CO₂ emissions from burning mattresses are the main reason for this discrepancy. Mattress recycling’s cost-effectiveness can be determined and expressed as an implied CO₂ price by comparing the costs and CO₂ emissions of the two options. According to our analysis, the cost of reducing one ton of CO₂ through mattress recycling is equivalent to 138 EUR/t CO₂.

As a point of comparison, the social costs of carbon dioxide in 2017 were approximately USD 46 (or EUR 50) for a ton of CO₂ emissions [16]. At the end of 2024, it is EUR 75 (or approximately USD 80). In addition, our implicit price is much higher than the current ETS price of EUR 75. Additionally, from a policy standpoint, it is critical to contrast it with viable options like wind energy, solar PV energy, or CCS. The cost of reducing one ton of CO₂ was shown to be EUR 30 for wind energy and EUR 81 for solar PV based on long-term calculations [2]. CCS in the North Sea, one of the more expensive energy options, costs between EUR 80 and 90 per ton of CO₂. This option still costs a lot less than our cover price for recycling plastic waste. In other words, compared to market and conventional CO₂ prices, as well as the available alternative technologies, the reduction in CO₂ emissions brought about by recycling plastic is expensive. In Figure 1 the costs in EUR per ton of CO₂ of the viable alternatives are given.

4.2. Sensitivity Analysis

We performed a sensitivity analysis based on a variety of plausible alternative assumptions to assess the reliability of our main conclusions. Utilizing four scenarios, the sensitivity of cost-effectiveness is evaluated. Following the baseline scenario,

Table 4. Cost of CO₂ reduction between mattress recycling and incineration scenarios.

	Cost Gap (EUR)	CO2 Emission Gap (t)	Cost of CO2 Reduction (EUR/t)
Base Scenario	224	1.62	138
Scenario 1: CO2-neutral replacement energy	224	2.39	94
Scenario 2: Treatment costs of incineration double	99	1.62	61
Scenario 3: EPR of EUR 10.47	123	1.62	76
Scenario 4: Proportion of 25% mattresses incinerated	245	1.35	181

displays the outcomes of the other scenarios. In the base scenario we assume that the opportunity CO₂ emissions as a results of recycling mattresses are based on the average energy mix in the Netherlands. The first scenario, therefore, like in [2], assumes that the replacement energy produced is renewable energy with no CO₂ emissions (i.e., fully CO₂ neutral). As a result, the associated opportunity CO₂ emissions are decreased from 0.76 to 0 tons. In this instance, energy recovery will result in lower CO₂ emissions than in the base scenario (0.45 instead of 1.21 per ton of mattress waste), increasing the emission difference while maintaining the same cost (see Table 4). In this case, the cost-effectiveness decreases to EUR 94 per ton of carbon dioxide.

The second scenario involves applying higher processing costs for mattress incineration. Several sources [29,32] suggest that the processing costs for incineration could turn out higher. The argument of Mulder (2018) is that mattresses have a higher calorific value than residual waste, and that other residual waste should be rejected if mattresses are incinerated [32]. Notably, in the long run, such a fixed supply of energy is not reasonable. To account for this, this analysis assumes an increase in processing costs of incineration from EUR 125 to EUR 250 per ton, and so the cost gap is reduced by EUR 125 (see Table 4). This reduces the cost effectiveness to 61 EUR/t CO₂.

In the third scenario we change the assumptions regarding the extended producer responsibility (EPR) as this is an important cost for mattress recycling. In the base scenario a fee of EUR 12.50 for a mattress is assumed according to [20]. Most recently, MRN reports a recycling fee of EUR 10.47 (including VAT), but according to the Royal Dutch association for waste management and cleaning (NVRD) this is not sufficient to reimburse all costs to municipalities [9]. In addition, increased transport costs are also mentioned. Nevertheless, we assume that on average the waste management or recycling fee will be reduced to a maximum of EUR 10.47 (as indicated by MRN, including VAT) and this decreases processing costs by EUR 41 and therefore increases the cost-effectiveness to 113 EUR/t CO₂.

Finally, a fourth scenario with a lower recycling rate than 90% is calculated. Particularly in the large cities such as Amsterdam, Rotterdam, the Hague, and Utrecht, we find that a substantial proportion of mattresses are contaminated or wet, which means they must be incinerated [33]. An expert consulted [30] indicated that in the Utrecht region, Rotterdam, and The Hague, 20% to 25% of separated mattresses are so heavily contaminated that they must still be incinerated. The assumption here is that the EPR is applied for all types of mattresses, so for both clean and contaminated mattresses. Incinerating a quarter of the mattresses will reduce the recycling rate from 90% to 75%. As a consequence, the recycling scenario is more costly and saves less CO₂. Therefore, the cost gap becomes EUR 245 and the CO₂ gap becomes 1.35 tons. This will increase the cost for CO₂ reduction, which will rise to 181 EUR/t, EUR 43 more than the baseline scenario.

5. Discussion

The results of this analysis have several policy implications. Although our analysis is based on Dutch data, the results are likely to be applicable to other countries given the composition of mattresses is similar in Europe and the USA. Moreover, in most European countries and US States mattresses are frequently landfilled, and incineration should not be seen as bad policy [12]. However, if incineration is to be avoided, as is increasingly the case in Dutch policy, it is important to reduce the cost of mattress recycling and to encourage innovation, for example, through advanced mechanical recycling [34]. Policies should encourage innovation in reducing or reusing mattress materials rather than focusing on the overly simplistic recycling target for different materials to increase innovation and thus opportunities for cost reduction. For example, the redesign of mattresses, including the extension of product life, is not stimulated by such a target. Although many innovations are still in development, the focus should be on stimulating a combination of high-quality reduce, reuse, and recycle activities [4] rather than an exclusive commitment to recycling. It is also important that citizens bring their mattresses to waste recycling points where they can be kept clean and dry, and this is a challenge in large cities. This is why it is important that manufacturers encourage people to take their old mattress with them when they receive a new one. However, as many people keep their old mattresses, for example, for guests, the question is whether this will yield enough and whether a strict recycling target is feasible for large cities. This cost-effectiveness analysis also shows that a further increase in the recycling fee to facilitate the collection of discarded mattresses by municipalities may be sub-optimal. This conclusion is like that for (packaging) glass, where the Dutch government wants to recycle more than the European target [5].

Let us discuss several issues which can be explored in future research. Firstly, our approach has focused on CO₂ or climate change, but recycling systems may also generate other life-cycle environmental impacts. [11] note that the literature comparing the life-cycle costs of incineration and recycling is sparse. This requires a full cost–benefit analysis, although there is no evidence to suggest opposite results. In general, recycling systems generate negative environmental impacts from the collection and processing of recyclable materials and positive environmental impacts when recycled materials displace virgin materials in production and processing: virgin materials vary widely depending on the type of material recycled. Secondly, as is common in a cost–benefit analysis, this analysis was based on current technology [2]. It would be worth investigating further whether improvements in incineration or recycling techniques are feasible. Garrido et al. (2016) and Serrano et al. (2024) show that pyrolysis coupled with the thermal or catalytic cracking of mattress PU foam waste in a laboratory-scale plant produces a rich gas with a high calorific value [13,35]. However, this process of pyrolysis needs a significant amount of energy. Liberati et al. (2024) advocates mechanical recycling by using air-lay techniques instead of rebounding PU foam from mattresses [34]. Moreover, such methods may increase the recycling costs or the price to the consumers and should be considered in a cost–benefit analysis. Thirdly, it is interesting to note that in the Netherlands the basis for mattress recycling is collectively managed by the industry itself. In future research it is worth investigating the price of the recycling fee, as higher than socially optimal prices may be the result. While a collective EPR offers benefits to municipalities and producers, it dilutes eco-design incentives and may raise competition concerns [19,36].

6. Conclusions

Compared to alternative viable technologies such as wind, solar, and CCS, as well as market-based and social CO₂ prices, the CO₂ reduction achieved by recycling Dutch mattresses is expensive. Using a cost–benefit analysis we show that the implicit cost of reducing one ton of CO₂ under the current mattress recycling program is EUR 138, which is significantly higher than ETS prices or alternatives such as wind energy. We show that there are two main reasons for this. First, the collection and processing costs of mattress recycling are high. Recycling is a complex process due to the heterogeneous composition of mattresses [10]. In addition, there are (modest) revenues from recycled mattress material waste, such as bonded PU foam and latex. Second, recycling mattresses has a relatively modest impact on CO₂ emissions compared to incineration. The typical Dutch household will separate about 0.9 kg of mattress waste yearly, saving 4.8 kg of CO₂ per year. The benefit of mattress recycling is very modest, accounting for less than 0.1% of total Dutch CO₂ emissions.

In addition, we show in a sensitivity analysis that the conclusion still holds when the main assumptions are relaxed. First, if energy production is CO₂ neutral then the cost-effectiveness improves from EUR 138 to EUR 94 per ton of CO₂—higher than the external costs reported in the literature. Second, if the treatment costs of incinerating mattresses are much higher and double, the cost-effectiveness improves from EUR 138 to EUR 61 per ton of CO₂, still higher than most other available alternatives. Third, if the recycling fee is lower at EUR 10.47, the cost-effectiveness improves from EUR 138 to EUR 76 per ton of CO₂—again higher than the external costs reported in the literature. Furthermore, if the separated mattresses are of good quality, 90% can be recycled. However, particularly in large cities, a significant proportion of mattresses are contaminated or wet, which means they must be incinerated. Fourth, based on empirical evidence for The Hague and Utrecht, if a quarter of the separated mattresses are incinerated then the cost-effectiveness becomes higher, changing from EUR 138 to EUR 181.